FR2723954A1 - PROCESS FOR THE PRODUCTION OF LINEAR SILOXANES OF LOW MOLECULAR MASS. - Google Patents

Abstract

The present invention relates to a process for producing low molecular weight linear siloxanes, which comprises the steps of: (A) mixing silicones containing hexamethyldisiloxane units with silicones containing di-organosiloxane units, (B) adding a catalytic amount of a soluble catalyst to the mixture of said step (A), (C) deactivating the rearrangement catalyst, and (D) collecting the linear low molecular weight siloxanes.

Description

Process for the production of linear low molecular weight siloxanes

  The present invention relates to a new process for the preparation of linear low molecular weight siloxanes. More particularly, it relates to the use of linear phosphonitrilic chloride type catalysts for rearranging M and D species to produce low molecular weight siloxanes such as

  octamethyltrisiloxane and decamethyltetrasiloxane.

  Octamethyltrisiloxane, also referred to as MDM, can be used effectively as a solvent in electronic device cleaning systems. It is known in the art to use a method of equilibration based on the use of Filtrol, to achieve equilibrium between MM and a source of D, so as to obtain as high a level of MDM as possible. the equilibration being followed by separation of the Filtrol by filtration and distillation of the product. The equilibrium MDM content is low and there are large amounts of MM and MDXM (x> 1). MM and higher boiling products can be recycled. Although the overall conversion rate from M and D to MDM is possibly high through recycling, the cost of the process per pound of product is also very high. In addition, the use of Filtrol and the need for filtration at each pass further reduces the overall efficiency of the process. The manufacturing scheme is so complicated and so time consuming that

  manufacturing cost becomes too high.

  In U.S. Serial No. 08/092450, filed July 15, 1993, under review, the contents of which are incorporated herein by reference, the applicant and his co-inventors disclose a novel process. which is characterized by a disproportionate reaction. They have found that if two or more polymers containing M and D and having different molecular weights are combined, in weight ratios of from about 1:99 to about 99: 1, preferably from about 5:99 to about at 95: 5, in the presence of a condensation / disproportionation catalyst such as a linear phosphonitrilic chloride, an extremely rapid and complete siloxyl disproportionation reaction occurs between the polymers containing M and D. The reaction results in the formation of a product having a lower molecular weight than one of the two raw materials,

  without formation of a significant amount of cyclic products.

  The applicant has pursued his research to develop a novel and simple process for large-scale production of low molecular weight siloxanes, such as octamethyltrisiloxane and

decamethyltetrasiloxane.

  In accordance with the present invention, there is provided a novel process for producing octamethyltrisiloxane and other linear low molecular weight siloxanes. The method comprises the step

  using a linear phosphonitrilic chloride (hereinafter referred to as

  after LPNC) to equilibrate a mixture of siloxanes comprising trimethylsiloxanes and dimethylsiloxanes, constituted

  preferably substantially hexamethyldisiloxanes (hereinafter

  after per MM units) and diorganosiloxanes (hereinafter referred to as D units). The LPNC catalyst makes it possible to obtain a very fast rearrangement of the siloxanes in the absence of silanol or of a quantity

  excessive water. The reaction does not require a filtration step.

  The present invention provides an efficient method for producing low molecular weight linear siloxanes having the general formula:

MDXM

  in which M represents trimethylsiloxane, D represents the

  dimethylsiloxane and x represents an integer greater than 0.

  The raw material used in the process of the present invention is a mixture of siloxanes containing MM units and D units. MM units and D units may be any of the existing silicone products, from liquid silicones to to silicone gums. The preferred ratio of MM units to

units D is 0.5: 1 to 10: 1.

  Among the catalysts which are preferred for use in carrying out the process according to the invention are phosphorus and nitrogen compounds which have been used in the prior art as condensation catalysts for the preparation of high molecular weight disilanols. Some of these catalysts are described in U.S. Patent Nos. 5,210,131, 4,902,813, 4,203,903 and 3,706,775, the contents of which are incorporated herein by reference. Examples of such phosphorus and nitrogen catalysts, useful as polycondensation catalysts and redistribution of organopolysiloxanes,

  include Cl3PN (PNCl2) xPC13.PC16 (x = 1) (LPNC) and phosphotes

  linear short-chain phazenes of formulas (Ia) and (Ib) O (X) 2 -mYmP {NP (X) 2} nNP (X) 3q (Y) q (Ia) O (X) 2-mYmP {NP (X) Where n is 0 or is an integer of 1 to equal to 0 or represents an integer equal to 1, q is equal to 0 or represents an integer equal to 1 or 2, X represents a halogen atom and Y represents an OH, OR or O (O) CR group, R representing a group

  alkyl or aryl. These examples are not limiting.

  Also useful catalysts include the reaction products of linear PNC compounds with compounds containing active protons having pKa values of less than 18, such as carboxylic acids, haloalkanecarboxylic acids, sulfonic acids, alcohols and alcohols. phenols. Cyclic PNC compounds, such as (PNCl 2) x, are also effective but act very slowly relative to linear catalysts. The compound Cl3PN (PNCl2) xPCl3PCl6 (x = 1) is the most preferred catalyst. The amount of catalyst used is an amount making it possible to easily achieve the disproportion of the siloxane system. The amount to be employed is not of fundamental importance and may vary from 5 to 500 parts, preferably from 10 to 100 parts, by

  million parts of total weight of organo-

  siloxanes used in the practice of the invention. A higher amount of catalyst will be used when the silanol content of the raw materials is high and is, for example, 1000 to 3000 ppm silanol. A quantity of PNC catalyst for condensing the raw material is first introduced into the reactor until the Si-OH content reaches a value of less than approximately 1000 ppm. A second addition of the

  PNC catalyst to achieve the disproportionation reaction.

  Preferably, the catalyst is dispersed or dissolved in an inert medium at a concentration of 0.1 to 10% by weight, preferably 0.5 to 5% by weight, in order to facilitate the handling of the catalyst and its dispersion. in the reaction mixture. Solvents suitable for the catalyst include ethers such as aliphatic ethers, aromatic compounds such as toluene and benzene, liquid siloxanes, chlorinated organic solvents, aliphatic or aromatic solvents, such as methylene chloride, trichloroethane, benzene and the like. 1,3,5-trichlorobenzene, etc. For carrying out the process according to the present invention, it is important that the silanol content of the organosiloxane mixture is low. In general, the upper limit of the content of

  silanol is approximately 3000 ppm relative to the total weight of

  siloxanes of the mixture. It is preferably 1000 ppm and even more preferably 700 ppm. For higher silanol contents, the silanol or condensation water hydrolyzes the LPNC catalyst. The hydrolyzed catalyst is a poor rearrangement catalyst. The problem can be solved by first adding a certain amount of LPNC and using the LPNC condensation action to reduce the silanol content to below 700 ppm, then to

  adding fresh LPNC to bring the rearrangement to a successful conclusion.

  The rearrangement reaction is continued until the desired equilibrium is reached and the reaction is then stopped. Since the boiling points of MM, MDM and MD2M are 100 C, 154 C and 195 C respectively, it is relatively easy to separate the desired product from the reaction mixture by distillation. When the reaction mixture is subjected to fractional distillation, a large amount of MM of the reaction mixture is distilled off prior to distillation of the product. Therefore, the catalyst must be deactivated at the end of the rearrangement and before the distillation. If the catalyst is not completely deactivated prior to the distillation of MM, the removal of MM in the presence of the active catalyst will result in a change in the equilibrium concentrations of the pot constituents as the MM content of the pot. reactor decreases. The catalyst can be chemically deactivated using any of a variety of bases. For example, the catalyst can be rendered inactive by neutralization with an alkaline product. Products

  Suitable alkalis include ammonia, hexamethyl-

  disilazane, aliphatic organic amines, primary, secondary or tertiary, such as ethylamine, diethylamine, triethylamine, propylamine, etc. The catalyst can also be chemically deactivated by means of inorganic products such as sodium hydroxide, calcium oxide and magnesium oxide. The amount of neutralizing agent used should be sufficient to stop any further rearrangement of M and D in the reaction mixture and provide a stable product. The amount is determined by reference to the total acid content and the amount is generally from 10 to 100 ppm neutralizing agent based on the total weight of the reagents. If the catalyst is supernatured by means of a base, an excessive amount of LPNC will be required to catalyze the successive batches again. According to one variant, the catalyst can also be thermally neutralized by raising the temperature to a value of C at 250 ° C. The temperature is fixed by the M / D ratio in the reactor. The higher the MM level, the lower the pot temperature. If the M / D ratio is too high, the reactor temperature is probably too low to allow thermal deactivation of the batch in a reasonable time. The reactor temperature can be raised to an appropriate deactivation temperature if the

  reaction system is pressurized.

  The distilled MM can be recycled to the reactor after separation of the MDM fraction and other low molecular weight linear species such as MD2M and MD3M, to be rebalanced with the higher boiling species that remain in the reactor. the reactor before the next distillation, fresh LPNC and species M and D being added also to the reactor for maintenance

  primitive M / D ratio and catalyst level.

  If the raw materials contain certain impurities, for example a high level of T units, after several recycling, these impurities can accumulate in the pot and reduce the efficiency of the process. The higher the level of T, the fewer the recurrences. At a certain point, when the level of T in the system is too high, the recycled residue stream must be purified separately. The rearrangement process may be a batch process or a continuous process. In a batch process, it is important to choose the optimum M / D ratio to ensure (1) a pot temperature adequate for catalyst deactivation in a short period of time, (2) a relatively low ratio of MM to Equilibrium MDM; and (3) a maximum amount of MDM in the reactor prior to distillation. It is possible that for the highest level of MDM the amount of MM to be distilled off as a "first cut" is also very high. Therefore, in a batch process, there is an optimum of MM and MDM which produces the highest extraction of MDM, MD2M, etc., per hour of process, and this optimum may not correspond to the maximum level of the

MDM content.

  In a continuous process, the M / D ratio may be less important because it is possible to design the apparatus for any level of MM at equilibrium, and the optimal method will probably correspond to the maximum MDM content. The fact that the equilibrium concentrations of MM and in particular MDM are approximately constant over a wide range of the M / D ratio, allows the distillation column portion of the process to work in the steady state even though the concentrations in the reactor vary. The M / D ratio at the input can be chosen to maximize the MDM, MD2M, etc., in the equilibrium mixture, and choose, for example, M / D ratios of 2/1 to 6/1 . Under these conditions, the ratio of MM to MDM at equilibrium will be high (about 1.5: 1 to about 3.0: 1), but this fact can be easily accounted for in the design of the column. All currents that are not product, can be recycled in a starting reactor. The starting reactor equilibrates the sources of M and D in the mixture, and the mixture is passed through a mixer through a static mixer at elevated temperature to deactivate the catalyst rapidly before the reaction. separation of MM and the fractions constituting the product in a distillation column operating continuously,

  specially designed for the separation of interesting products.

  A continuous process for higher volume production may require two columns. An equilibrium mixture, deactivated, is introduced into the first column where either 1) MM is removed from the top, 90% MDM is removed by the medium and heavy fractions are removed from the bottom, ie 2) MM. and MDM (may be 50% MDM) are removed together at the top and heavy fractions are removed from below. The second column carries out the final purification. It is best to make the first column a part of the continuous system and the purification

  final is carried out discontinuously.

  The process can be carried out in a simple column / reboiler reactor, with many recycles, by recycling the MM from the first distillation cup directly into the reboiler, with some M and D booster corresponding to the product being removed. From the point of view of cost, this method has considerable advantages over prior art methods. The silicones produced by this simple, fast, low-cost, low-by-product process have many uses, for example as solvents which can

  replace halogenated solvents in the electronics industry.

  The following examples are given for purposes of illustration of the present invention. All viscosities are measured at 25 C. Unless otherwise stated, all parts are mole parts.

EXAMPLE 1

  The equilibrium distribution of the various MDXM species (x0O) is determined for various values of the M / D ratio. Appropriate quantities of MM and liquid silicones having a viscosity of 350 cps are weighed into flasks to obtain 20 g of mixture for each value of the M / D ratio. The LPNC catalyst is available under the

  form of a 2% solution in a dimethylsilicone oil (SF-

  96 (20)). 100 ppm of LPNC catalyst (0.1 g of the catalyst solution) is introduced into each flask. This level of catalyst is much higher than necessary. Rearrangement already occurs with 20-25 ppm LPNC. An excess amount is used to ensure that the experiments run smoothly. Since the catalyst solution is primarily a D silicone, the 0.1 g of catalyst solution is included as part of component D for the preparation of M / D ratios, so as to avoid

  compositional errors at high M / D ratios.

  The flasks are placed in a boiling water bath for 3 hours and then cooled to room temperature. 10 ml of HMDZ are introduced into the flasks and reacted with all the chloride present in the LPNC. Solutions become troublesome

after a few minutes.

  All samples are analyzed using a HewlettPackard 5890 GC chromatograph with a 30-meter capillary column and thermal conductivity detector. Gas chromatographic analysis gives a percentage of area for each constituent. The retention factors of the various constituents

  are not determined and only percentages of area are indicated.

Table 1 *

A B C D E F

  REPORTM / D 0,5 / 1 1/1 2/1 4/1 6/1 10/1

MM 11 21 37 51.6 60 68

MDM 13 20 24 25.8 25 23

MD2M 12.6 16.4 16.7 13.6 9.2 5

MD3M 11.4 12.1 9.0 4.8 2.3 0.9

MD4M 9.4 8.6 4.7 1.6 0.6 0.2

MD5M 9.4 6.0 2.5 0.5 0.2 -

MD6M 7.1 4.2 1.3 0.17 -

MD7M 5.8 2.9 0.7 0.1 -

MD8M 4.8 1.9 0.4 -

MD9M 3.8 1.3 0.23 - -

D4 3.0 1.3 0.4 0.15 0.12

  * Small peaks (<0.1%) are not indicated.

  The data show that as the M / D ratio increases, there is a steady shift towards linear species of lower molecular weight and a greater amount of equilibrium MM. It has now been surprisingly found that the percentage of MDM varies only slightly (20 to 26%) when the M / D ratio varies from 1: 1 to 10: 1. This relative insensitivity of the MDM content to the variations of the M / D ratio in this area is particularly advantageous for the adjustment of a continuous process. For an M / D ratio in the range of 1: 1 to 4: 1, the percentage of

  component MD2M also varies little (from 13.6% to 16.7%).

  Figure 2 is a graph showing the percentages of MM and MDM in each equilibrium mixture, as a function of the M / D ratio. This graph is constructed to show the percentage of the lot corresponding to MM and MDM, and the ratio of MM

to MDM, based on the M / D ratio.

Table 2

  RatioM / D% MM% MDM% of (MM + MDM) MM / MDM 0.5 / 1 1il 13 24 0.85

1/1 21 20 41 1.05

2/1 37 24 61 1.54

4/1 51.5 25.8 77.3 2.0

6/1 60 25 85 2.4

/ 1 68 23 91 2.95

  The results clearly show the relative insensitivity of the MDM and MD2M contents to the M / D variations. This is because a pound of D produces 3.18 pounds of MDM, which is a major advantage for the conversion of what appears to be a considerable excess of M to MDM at high M / D ratios. The equilibrium MM content is only doubled when the ratio

M / D goes from 2/1 to 10/1.

  However, Figure 3 is a graph showing the ratio of MM to MDM as a function of M / D, which shows increasing inefficiency

  distillation as the M / D ratio increases.

  A 1: 1 M / D ratio requires the distillation of only 1 pound of MM per pound of MDM, but these two species account for only 41% of the pot content at this M / D value, and therefore much more is needed. batch reactor tests per pound of

produced at the exit.

EXAMPLE 2

  In an attempt to determine the optimum conditions for a batch process, MM and the liquid silicone are introduced into a flask at a molar ratio M / D equal to 1: 1 and the system is refluxed. The reflux pot temperature is 104 ° C. 100 ppm LPNC is added. The temperature rises regularly to 149 C in the space of 2 hours. The charge is kept under reflux for a further 2 hours to ensure thermal deactivation of the catalyst. At this point, additional MM is added to the flask so that the M / D ratio is 2/1. The pot temperature 1l stabilizes at 130 C. 100 ppm additional LPNC is introduced into the flask. The pot temperature rises to 138 ° C. This shows that the initial reaction of 2 hours at 149 ° C. is suitable for the complete deactivation of the catalyst. The result also shows that the rearrangement starts easily after

  the addition of recycled MM and more LPNC to the load.

  As shown in Table 2, for a load of 10M pounds, it is necessary to distil only about 4M pounds of feed to produce about 2M pounds of MDM and, if desired, to produce 1.6M pounds of MD2M. For an M / D ratio of 4/1, it would be necessary to distil 7.7M pounds to produce 2.4M pounds of product. Load balancing / turnarounds seem easy with the LPNC and as the heat-off time would be long at the equilibrium pot temperature, in the case of an M / D ratio greater than or equal to 2/1, the High M / D ratios are not favored. When the M / D ratio is greater than 2/1, the equilibrium pot temperature can be raised to an appropriate deactivation temperature if the reaction system is pressurized. Although particular examples of the process according to the invention have been described above, the invention is not limited to these examples alone, but encompasses all variations and modifications.

  within the scope of the appended claims.

Claims (20)

  A process for producing linear siloxanes of low molecular weight, characterized in that it comprises the steps of:   (A) mixing silicones containing hexamethyl-   disiloxanes with silicones containing di-   organosiloxanes, (B) adding a catalytic amount of a soluble catalyst to the mixture of said step (A), (C) deactivating the rearrangement catalyst, and   (D) recover linear siloxanes of low molecular weight.   2. Method according to claim 1, characterized in that the   preferred ratio of hexamethyldisiloxane units to diorgano siloxanes is 0.5: 1 to 10: 1.   3. Process according to claim 1, characterized in that the catalyst comprises a phosphonitrilic chloride and / or a chloride oxygenated phosphonitrilic acid.   4. Method according to claim 1, characterized in that the catalyst comprises a phosphonitrilic halide catalyst having the general formula [X (PX2 = N) nPX3] + [MX (Vt + I) R2t] - in which X represents an atom of halogen, M represents an element having an electro-negativity of 1.0 to 2.0 on the Pauling scale, R2 represents an alkyl group having up to 12 carbon atoms, n is an integer having a value of from 1 to 8, v is the valence or degree oxidation of M and 0 <t <v.   5. Method according to claim 1, characterized in that the   The catalyst comprises carboxylic acids, halogeno   alkanecarboxylic acids, sulfonic acids, alcohols or phenols. 6. Method according to claim 1, characterized in that the content   in silanol of said siloxane mixture is less than 3000 ppm.   7. Method according to claim 6, characterized in that the content   in silanol of said siloxane mixture is less than 1000 ppm.   8. Process according to claim 7, characterized in that the content   in silanol of said siloxane mixture is less than 700 ppm.   9. Process according to claim 1, characterized in that the   Rearrangement catalyst is thermally deactivated.   10. Process according to claim 9, characterized in that the rearrangement catalyst is thermally deactivated at a temperature of   temperature between 130 C and 250 C.   11. Process according to claim 1, characterized in that the   The catalyst is rendered inactive by neutralization with an alkaline product.   12. Process according to claim 11, characterized in that the alkaline product is chosen from ammonia, hexamethyldisilazane and the primary, secondary or tertiary aliphatic organic amines, such as ethylamine, diethylamine and triethylamine. propylamine.   13. Process according to claim 1, characterized in that the   The catalyst is rendered inactive by neutralization with a mineral product.   14. Process according to claim 13, characterized in that the mineral product is chosen from sodium hydroxide, calcium and magnesium oxide.   15. Process according to claim 1, characterized in that the siloxanes of low molecular weight correspond to the general formula: MDXM   in which M represents trimethylsiloxane, D represents the   dimethylsiloxane and x is an integer greater than 0.   16. The method of claim 15, characterized in that the   low molecular weight siloxane is octamethyltrisiloxane.   17. The method of claim 15, characterized in that the   Low molecular weight siloxane is decamethyltetrasiloxane.   18. The method according to claim 1, characterized in that said step (D) comprises the step of collecting the siloxane of low   molecular mass by distillation.   19. The method of claim 1, characterized in that said step (D) further comprises the step of collecting   hexamethyldisiloxane by distillation.   20. The method of claim 19, characterized in that it further comprises the step (E) of recycling hexamethyldisiloxane.